Understanding losses in three core armoured submarine cables

As the number of offshore wind farms increases so does the number of array cables. The array cables are used to interconnect the wind turbines in a wind farm and to connect the wind farm to a collector platform, whereas the connection to shore is performed by means of a HV transmission cable. For practical an economical reasons the preferred choice of cable for both the array and the transmission cables are three-core armoured submarine cables. Therefore, it has becoming increasingly important to be able to calculate the ampacity of such cables accurately. At present time, the ampacity of three-core armoured submarine cables is calculated according to IEC 60287-1-1 [1]. Various measurements conducted both by cable manufacturers and transmission system operators (TSO) have shown that using the cable rating method stated in IEC 60287-1-1 underestimates the cable ampacity [2]-[6]. Furthermore, measurements conducted within the cable industry have shown that an armoured three core cable has higher losses than equal unarmoured three core cables. It is also suggested that the inaccuracy in the IEC armour’s loss factor (λ2) is the main responsible for the conservatism in the IEC cable rating method. However, research performed in this paper shows that while the armour is responsible for the extra losses, the main calculation error for typical HVAC submarine cables may be in the screen’s loss factor (λ1).
Overestimation of the cable losses may result in core cross-sections too large with more material than required and thereby a more costly cable, as well as more expensive installation. Therefore, further research is needed in order to develop new analytical equations capable of a more accurate estimation of the losses in three-phase armoured cables. The nature of the work presented in this paper is to investigate the influence of different parts of three-core armoured cables and of the respective electrical parameters, resistivity and permeability, in the total losses. Different cables are simulated via Finite Element Methods software and the results are compared with those obtained using IEC 60287-1-1, together with theoretical analysis of the results. It is found that formulas capable of accurately estimate the screen’s loss factor already exist in the literature, but the formula for estimation of the current in the screen, a parameter necessary in order to estimate the loss factor, needs improvements.